Biomedical Engineering Reference
In-Depth Information
Thus, reliability means that something will not fail prematurely. Or, stated more positively, reliability is
expressed mathematically as the probability of success. Thus reliability is the probability that something
that is in operation at time 0 (t
0
) will still be operating until the designed life [time
t
t
]. People who
will receive an implanted device want to be assured that it will work and will not fail. If t
t
=
=
100 years,
most patients would be happy. If t
t
=
30 days for absorbable sutures, this may be acceptable, but very
likely not to be acceptable for nonabsorbable sutures (depending on the actual use by the physician).
Thus, “failure” depends on the expected life of the engineered system.
The probability of a failure per unit time is the “hazard” rate, a term familiar to risk assessment, but
many engineers may recognize it as a “failure density,” or
f(t)
. This is a function of the likelihood that
an adverse outcome will occur, but note that it is not a function of the severity of the outcome. The
f(t)
is not affected by whether the outcome is very severe (e.g., complete failure to provide oxygen) or
relatively benign (e.g., muscle soreness or very small amount of leakage of a nontoxic lubricant). It is
up to the designer to decide which factors are to be tested for reliability.
The likelihood that something will fail at a given time interval can be found by integrating the hazard
rate over a defined time interval:
t
2
Pt
1
≤
T
f
≤
t
2
=
ftdt
(5.1)
t
1
where T
f
=
time of failure.
Thus, the reliability function
R(t)
of a system at time
t
is the cumulative probability that the system
has not failed in the time interval from t
0
to t
t
:
t
Rt
=
PT
f
≥
t
=
1
−
fxdx
(5.2)
0
One major point worth noting from the reliability equations is that everything we design will fail.
Engineers can improve reliability by extending the time (increasing t
t
). And, this is done by making the
system more resistant to failure. For example, proper engineering design of a barrier between tissues
can decrease the migration of a fluid. However, the barrier does not completely eliminate failure, i.e.,
Rt
0; it simply protracts the time before the failure occurs (increases T
f
). Reliability can be affected
by societal factors. For example, health studies in much of the twentieth century focused on adult white
males to a great extent. It is possible to expect a decrease in T
f
because certain attributes of children,
women, and minorities were not properly considered in the original design of a device (e.g., the presence
of different chemicals in the endocrine, neural, and immune systems, which changes the performance
of a device).
A discipline within engineering, i.e., reliability engineering, looks at the expected or actual reliability
of a process, system, or piece of equipment to identify the actions needed to reduce failures, and once
a failure occurs, how to manage the expected effects from that failure. Thus, reliability is the mirror
image of failure. Since risk is really the probability of failure (i.e., the probability that our system,
process, or equipment will fail), risk and reliability are two sides of the same coin. A device leaking
chemicals into the bloodsteam is an engineering failure, as is exposure of people to medical monitoring
and sensing that has not been properly tested. A system that protects one group of people at the expense
=
